Bacterial Cyclopropane Fatty Acid Synthase mRNA Is Targeted by Activating and Repressing Small RNAs

Small RNAs (sRNAs) in bacteria are abundant and play important roles in posttranscriptional regulation of gene expression, particularly under stress conditions. Some mRNAs are targets for regulation by multiple sRNAs, each responding to different environmental signals. Uncovering the regulatory mechanisms governing sRNA-mRNA interactions and the relevant conditions for these interactions is an ongoing challenge. In this study, we discovered that multiple sRNAs control membrane lipid composition by regulating stability of a single mRNA target. The sRNA-dependent regulation occurred in response to changing pH and was important for cell viability under acid stress conditions. This work reveals yet another aspect of bacterial physiology controlled at the posttranscriptional level by sRNA regulators.

: Deletion of the putative sRNA binding site removes regulation of cfa translation by ArrS, CpxQ, and RydC (A) 5! UTR of cfa gene. Arrows mark transcriptional start sites and sRNA binding sites are indicated with labeled boxes.
(B) A cfa translational fusion to lacZ (cfa'-'lacZ-ΔsRNABS) that begins immediately downstream of the predicted sRNA binding sites was constructed. This fusion contains the proximal σ s -dependent promoter and 79-nt upstream of this promoter. Regulation of cfa'-'lacZ-ΔsRNABS by ArrS, CpxQ, and RydC was tested as described in Figure 1B. Figure S3: Deletion of rydC does not alter cfa mRNA regulation by CpxQ CpxQ mediated repression of cfa'-'lacZ-Long was tested in a WT or ΔcpxQ ΔrydC background as described in Figure 1B. The statistical significance was determined using a two-tailed Student's t-test, ns is for not significant.

Figure S4: The cfa 5' UTR differs between E. coli and Salmonella
Sequence alignment of the promoter and 5' UTR of cfa in E. coli and Salmonella. Gray box: Hfq binding site. Orange box: CpxQ binding site. Green box: Activating sRNA binding site. Black arrow: RNase E cleavage site. The transcription start sites are labeled. H1 and H4 mutations in the Hfq binding site are shown in a blue shaded box. G1, G2, and G3 mutations in the RNase E recognition site are shown in a yellow shaded box. Figure S5: Structure of WT and G3 mutant cfa mRNA 5! UTR (A) Secondary stucture of the WT cfa mRNA 5! UTR and (B) the G3 mutation in the cfa mRNA 5! UTR as predicted by mFold. The three nucleotides mutated in the G3 mutant are in a red shaded box. The CpxQ binding site (orange) and the RydC binding site (green) are labeled. Note that the structures of the CpxQ and RydC binding sites are the same in G3 mutant as in the WT. Numbered nucleotides are relative to the translational start site (labeled +1).

Figure S6
: DNA-oligo annealing to Hfq binding region does not alter sRNA binding (A) Secondary stucture of the cfa mRNA 5! UTR as predicted by mFold. The CpxQ binding site (orange) and the RydC binding site (green) are labeled. Numbered nucleotides are relative to the translational start site (labeled +1).
(B) In vitro structure probing using 5! end-labeled cfa mRNA with RNase T1 (lanes 4-9) and lead(II) acetate (lanes 10-15) in the presence of a ssDNA-oligo (oligo H1) that base-pairs to the Hfq binding region (-195 to -161) in the 5! UTR of cfa mRNA and either CpxQ or RydC. RNase T1 and alkaline ladders of cfa mRNA were used to map cleaved fragments. Positions of G-residues are indicated relative to the translational start site. The RydC, CpxQ, and oligo H1 base-pairing sites are labeled on the right. Figure S7: Deletion of sRNAs do not affect cfa'-'lacZ-Short activity (A) Cells carrying cfa'-'lacZ-Short were grown and cfa translation in response to acid was assayed as described in Fig. 8. The error bars are standard deviations of three biological replicates and the statistical significance was determined using a two-tailed Student's t-test, ns means not significant.
(B) Cells carrying cfa'-'lacZ-ΔsRNABS were grown and cfa translation in response to acid was assayed as described in Figs. 8, S7A.
(C) Cells carrying cfa'-'lacZ-Short (in either WT background or a background where one sRNA is deleted) were grown as described in Figs. 8, S7A. The error bars are standard deviations of three biological replicates and the statistical significance was determined using a two-tailed Student's t-test, ns means not significant.
(D) Cells carrying cfa'-'lacZ-Short (in either WT background or a background where rydC and one other sRNA are deleted) were grown as described in Figs. 8, S7A. ΔrydC single mutant is included for reference. Error bars represent standard deviation for three biological replicates and the statistical significance was determined using a two-tailed Student's t-test, ns means not significant.

Figure S8
: CpxQ expression is higher at pH 7 compared to at pH 5 Expression levels of CpxQ at pH 7 and pH 5 were determined by northern blot analysis of total RNA samples. Wild-type cells were grown in medium at pH 7 then subcultured into medium either pH 7 (three biological replicates) or pH 5 (three biological replicates). Samples were harvested 120 minutes later and RNA was extracted. The P LlacO CpxQ positive control was prepared as described in Fig. 4C. 5S served as loading control.
! sRNA  : Plasmids and strains used in this study  : Oligos used in this study Table S4